Subido por Cristian Melendres Orellana

BOOMERANG

Anuncio
Marine and Petroleum Geology 116 (2020) 104299
Contents lists available at ScienceDirect
Marine and Petroleum Geology
journal homepage: www.elsevier.com/locate/marpetgeo
Research paper
Kinematic and mechanic evolution of the Andes buried front at the
Boomerang Hills hydrocarbon province (Bolivia): Control exerted by the
pre-Andean history
T
Hodei Uzkedaa,∗, Mayte Bulnesa, Josep Pobleta, Gonzalo Zamorab
a
b
Departamento de Geología, Universidad de Oviedo, C/Jesús Arias de Velasco s/n, Oviedo, 33005, Spain
Repsol Exploración S.A., C/Méndez Álvaro 44, Madrid, 28045, Spain
ARTICLE INFO
ABSTRACT
Keywords:
3D geological model
Andes
Boomerang hills
Inversion tectonics
Sequential restoration
The Boomerang Hills are a hydrocarbon-rich region located in front of the Bolivian Andes, central South
America, where the Andean chain bends forming the Bolivian Orocline. The geological interpretation of seismic
(2D profiles and 3D volumes) and well data allowed us to generate a complete stratigraphic section, cross
sections, maps and a 3D model, and to obtain the parameters to carry out a depth conversion and decompaction.
A cross section and a 3D model were kinematically and mechanically restored, as well as decompacted, to
decipher the evolution of this region through time. The southern Boomerang Hills are the buried frontal part of
the Andes where thin-skinned contractional structures of Andean age (Late Cenozoic) predominate, whereas the
northern Boomerang Hills belong to the Beni-Chaco Plain, where contractional structures are almost absent and
thick-skinned extensional tectonics of pre-Andean age (Paleozoic, Mesozoic and Early Cenozoic) dominate. The
buried frontal thrust of the Andes resulted from transpressional reactivation of a gravity-driven, extensional
detachment developed along the slope of a basement high tilted during Paleozoic times. Thus, the thrust orientation, position, dimensions and type of structural termination were controlled by the extensional fault
geometry, the Paleozoic mechanical stratigraphy, and the structural relationships between the Paleozoic cover
and the basement.
1. Introduction
The Boomerang Hills constitutes one of the main prospective areas
in terms of hydrocarbons in central Bolivia (Fig. 1) and has been extensively explored from the sixties. Several 2D and 3D seismic surveys
have been acquired and hundreds of wells have been drilled in this
area, and more than 250 have tested hydrocarbons. Unraveling the
geology of the Boomerang Hills using surface data is unsatisfactory
because only the upper part of the Cenozoic succession crops out, thus
the good seismic and well coverage provides an excellent dataset to do
so. In addition to its economic interest, this region is remarkable from a
structural point of view because of two reasons: 1) a combination of
extensional, contractional and inversion tectonics structures, due to
different tectonic events, configure this area; and 2) the general trend of
the Andean Cordillera bends from NW-SE to the northwest, to WNWESE near the Boomerang Hills, to N–S to the southeast (Fig. 1) as part of
the so-called Bolivian Orocline (e.g. Carey, 1958; Isacks, 1988). The
most significant stratigraphic feature of the Boomerang Hills is a
∗
pervasive angular unconformity at the base of the Mesozoic sequence
that truncates progressively older Paleozoic rocks towards the north.
Although some geological issues have been addressed in previous
publications such as Baby et al. (1994, 1995), Welsink et al. (1995a),
Laffite et al. (1998), Kley (1999), Giraudo and Limachi (2001), Hinsch
et al. (2002, 2003), Husson and Moretti (2002), there are still important
aspects to understand regarding the stratigraphy and structure of the
Boomerang Hills as, for instance, the control exerted by previous structures and stratigraphy on the Andean deformation. Baby et al. (1995)
analyzed the source rocks and carried out petroleum modeling in order
to decipher when they entered the hydrocarbon generation window.
Welsink et al. (1995a) described the geology using seismic lines and well
data, and defined two structural domains based on the most common
structures and the stratigraphic sequence. Laffite et al. (1998) analyzed
the main hydrocarbon parameters, such as maturation, TOC (Total Organic Carbon), TAI (Thermal Alteration Index) and Ro (vitrinite reflectance) values. Kley (1999) constructed a kinematic model of the
Bolivian Orocline (from Peru to Chile) using restored sections and
Corresponding author.
E-mail addresses: [email protected] (H. Uzkeda), [email protected] (M. Bulnes), [email protected] (J. Poblet), [email protected] (G. Zamora).
https://doi.org/10.1016/j.marpetgeo.2020.104299
Received 4 October 2019; Received in revised form 11 February 2020; Accepted 13 February 2020
Available online 19 February 2020
0264-8172/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
2. Geographical and geological setting
The studied area, which covers an extension of about 7000 km2, is
the Boomerang Hills, central Bolivia. From the geological point of view,
this area is located in the Beni-Chaco Plain, i.e. the foremost zone of the
compressive Andean belt (Fig. 1). To the southeast of the Boomerang
Hills the Andes bend forming a concave arch towards the southwest,
whereas to the west of the Boomerang Hills the Andes form a convex
arch towards the southwest. However, the main structures of the
Boomerang Hills do not follow any of these directions, instead, they
strike approximately W-E. This orientation seems to be controlled by
the inherited basement configuration and the distribution of the cover
(Silurian to Cenozoic) sequence (Baby et al., 1994; Welsink et al.,
1995a; Giraudo and Limachi, 2001; Hinsch et al., 2003). Moreover, a
transition from a southern domain, with dominant shortening of Andean age, to a northern domain, with preserved extensional structures
of pre-Andean age, has been documented in this region (Welsink et al.,
1995a).
The Boomerang Hills have undergone several tectonic episodes from
Paleozoic to recent times. As a consequence, structures developed in
extensional and contractional settings, as well as some structures that
exhibit evidence of successive reactivations, coexist. The deposition of a
thick marine sequence of Cambrian-Ordovician age which, together
with Precambrian rocks configure the basement (Brazilian Shield), was
interrupted in late Ordovician times causing an unconformity in between the Ordovician and Silurian (Lindquist, 1998). During the Silurian, a rifting took place, inducing the establishment of anoxic marine
conditions favorable to the deposition of shaly source rocks (Lindquist,
1998). Several kilometers of marine sediments accumulated during the
Devonian, such as Devonian shale source rocks, as well as sandy beds
that may act as reservoirs. The sedimentation continued with Carboniferous glacial deposits formed by alternations of shaly and sandy beds
that can be source and reservoir rocks respectively (Baby et al., 1995).
Permian-Triassic rocks are absent, as well as part of the Paleozoic
succession. Thus, the middle-upper Mesozoic succession lays unconformably over Paleozoic rocks, and it is mainly formed by shales
with interbedded siltstones and sandstones deposited under continental
conditions (Welsink et al., 1995a). New normal faults developed during
Mesozoic and older ones were reactivated under this extensional regime. During Cenozoic times, conglomerates, sandstones and shales
were deposited in continental environments. Although a few normal
faults were generated and old ones were reactivated at the beginning of
the Cenozoic, the tectonic setting changed radically with the initiation
of the Andean orogeny and the establishment of a contractional regime
responsible for thrusting and reverse reactivation of older normal faults
(Welsink et al., 1995a).
Fig. 1. Tectonic map illustrating the main structural units surrounding the
Boomerang Hills (modified from Geobolivia, 2000). Red rectangle: study area
displayed in Figs. 2, 4, 5, 18 and 19. (For interpretation of the references to
color in this figure legend, the reader is referred to the Web version of this
article.)
paleomagnetic rotation data, and concluded that the motion component
perpendicular to the orogen increases in the region near the Andes bend,
whereas the parallel component increases as one moves away from it.
Baby et al. (1994) and Giraudo and Limachi (2001) determined how the
basement configuration and the Paleozoic sequence wedge affected the
structure development, and they pointed out that the absence of a sufficiently thick Paleozoic sequence induced the Andes bending. Hinsch
et al. (2002, 2003) studied how the obliquity of the basement with respect to the general shortening direction affected the position of the
deformation front and caused both a deformation partition and a reorganization of the main stress direction. They also showed the importance of the old basement faults and other irregularities on the nucleation of new faults. Husson and Moretti (2002) characterized the
thermal regime along the Bolivian Andes and their spatial variations,
determining an increase towards the Boomerang region. According to
these authors active tectonics in the Bolivian Subandean region causes
important variations in the regional thermal field. Finally, Moretti et al.
(2002) modeled the possible fluid migration and the influence exerted by
the porosity and textural characteristics of the rocks and other factors,
such as the thickness and features of the fault zones.
Understanding the main characteristics of the structures, their distribution and their evolution through time, as well as the distribution of
the different stratigraphic units and their thickness variations with
time, were the first aims of the present work. This allowed us to gain
insight into the influence exerted by the older stratigraphic successions
and structures on the features of the different types of structures developed during younger events, both at individual structure scale and at
the scale of this portion of the Andean Cordillera. In addition, it allowed
us to visualize the distribution of the findings of different types of hydrocarbons in relation to the different types of structures and their location within the region. In order to achieve these objectives, time
seismic profiles and 3D volumes were geologically interpreted with the
constrain of well data, and subsequently they were used to create
geological cross-sections, maps, a structural map and a 3D model. A
synthetic cross section, constructed by projecting the closest geological
interpretations onto a chosen plane, maps and the 3D model were
depth-converted, sequentially decompacted using functions created
specifically for the stratigraphic units present in this region, and sequentially restored using kinematic and mechanical algorithms.
3. Available data
The available data consists of two 2D seismic surveys, the Sara and
Palacios Norte, including respectively 213 and 9 time pre-stack seismic
lines, two 3D time seismic volumes (Sara Bomerang and Los Penocos)
and 31 wells distributed in 18 clusters (Fig. 2). These subsurface data
were acquired by the oil companies Yacimientos Petrolíferos Fiscales
Bolivianos (YPFB)-Andina and Maxus between the 60's and 2000's, who
made many hydrocarbon discoveries.
The Sara seismic survey covers the whole study area, whereas the
Palacios Norte seismic survey covers the central portion of the study
area. Most seismic lines display a N–S or a NE-SW trend, although a few
E-W and scarce NW-SE are also available (Fig. 2). Commonly, they have
a record length of 4000–5000 ms. The total length of all these seismic
lines is approximately 3900 km.
The 3D seismic volume called Sara Boomerang, located in the central-southwest part of the study area, is elongated in ENE-WSW direction and recorded 6000 ms (Fig. 2). The 3D seismic volume called Los
Penocos, located in the northeast sector of the study area, is elongated
2
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 2. Distribution of the available seismic surveys (lines and volumes) and wells superimposed on a structural map of the study area (see location of the study area
in Fig. 1). The map shows the intersection between the topographic surface and the structural elements of outcropping structures, as well as the intersection between
the topographic surface and the vertical projections of the structural elements of buried structures. The geographical coordinates and hydrocarbon contents of the
wells have been extracted from the IHS Markit Well Database (https://ihsmarkit.com/products/international-well-data.html). The names of the available wells are:
ICH: Ichoa, VRB: Víbora, CCB: Cascabel, SIR: Sirari, BQN: Boquerón, YPC: Yapacani, YAPA: Yapa, PJT: Patuju, PPL: Puerto Palos, PLC: Palacios, ECD: Enconada, LPS:
Los Penocos, SJN: San Juan, SGD: Santo Domingo, PNW: Palometas noroeste, and CTO: Cuatro Ojos. Víbora, Cascabel, Palacios, Enconada, San Juan and Palometas
correspond to the names of the drilled anticlines. The location of Figs. 6, 7, 9 and 13 is shown.
in ESE-WNW direction and recorded 2500 ms (Fig. 2). Both 3D volumes
cover an area of 630 km2. A total of 93 inlines and crosslines belonging
to the two 3D seismic volumes were interpreted from the geological
point of view, which makes a total of approximately 800 km. The
quality of the seismic imaging of 3D volumes is similar to that of 2D
seismic lines.
Most available wells are placed in the central-southern part of the
study area (Fig. 2). Wells are much scarcer to the north, except for a
couple of wells located in the area covered by Los Penocos 3D seismic
volume. The depth of the wells ranges between 1.7 and 1.8 km at SDGx1 and LPS-1 to LPS-4, around 4 km at SCR-x1, VBR-x1, SIR-x1 and ICHx1, and reaches almost 5 km at SJNx2 (south). Wells provided fundamental information in order to identify stratigraphic units within the
seismic data, to obtain a velocity model using interval velocities, to
carry out the depth conversion and to estimate the parameters necessary for decompacting the stratigraphic succession.
Three main geological elements have been interpreted on the time
seismic lines and 3D volumes using the software Kingdom Suite© and
Petrel©: tops of selected seismic-stratigraphic units, main faults and
main fold axial surfaces. The correlation between the stratigraphic
horizons drilled in the wells and the reflectors in the seismic data is
based on three types of data included in the confidential well reports
elaborated by the company YPFB-Andina: 1) time versus depth tables
and/or functions, 2) time values of the different stratigraphic horizons
drilled, and/or 3) zooms of interpreted seismic lines in which the wells,
along with the drilled stratigraphic horizons, are shown.
divided here in five mega-sequences bounded by major unconformities,
which from top to bottom are: Cenozoic, Mesozoic, Carboniferous,
Silurian-Devonian and pre-Silurian (modified from Welsink et al.,
1995a) (Fig. 3). The Cenozoic, Mesozoic, Carboniferous and SilurianDevonian successions are cover rocks, whereas the pre-Silurian succession is considered basement. Using the available well data, up to
eleven stratigraphic units were identified and used to carry out the
geological interpretation of the seismic data: Chaco, Yecua and Petaca
fms. (Cenozoic); Cajones, Yantata and Ichoa fms. (Mesozoic); Carboniferous; Iquiri, Limoncito, Roboré and Boomerang fms. (Devonian);
Kirusillas Fm. (Silurian); and pre-Silurian (usually considered Precambrian-Ordovician).
A brief description of the lithologies included in these stratigraphic
units, based on well data, is given below from younger to older units.
The main seismic-stratigraphic features are described in Welsink et al.
(1995a). The Cenozoic mega-sequence is formed by three continental,
siliciclastic formations. The Chaco Fm. (Cenozoic) consists of an alternation of sandstones and shalier units; sandstones predominate in the
upper part, and limolites and shales are the main constituent of the
lower part. The Yecua Fm. (Cenozoic) is made up of claystones with
sporadic sandy intercalations more frequent towards its base. The Petaca Fm. (Cenozoic) is constituted by sandstones and shales in variable
proportions, although in general, the upper part is sandier, whereas
shales dominate the lower section. Three continental-shallow marine,
siliciclastic formations, including chert and gypsum horizons, constitute
the Mesozoic mega-sequence. The Cajones Fm. (Cretaceous) is a fine to
coarse sandy unit with intercalations of shalier intervals as well as
calcarenites. This unit appears sporadically in the study area, which
prevented its tracking on the seismic data and obtaining a surface useful
for restoration and decompaction. The Yantata Fm. (Cretaceous) is
formed by sandstones with less shaly intercalations than the units
above, although they may be relatively important towards its top; chert
4. Stratigraphy
4.1. Stratigraphic units
The stratigraphic succession of the Boomerang Hills has been
3
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 3. Simplified stratigraphy of the Boomerang Hills and surrounding areas showing the geometry of the different units, their ages and the location of the potential
source rocks, reservoir rocks and seal rocks (modified from Baby et al., 1995). The angle between the Mesozoic and Paleozoic rocks has been exaggerated for better
visualization. Black square: study area, wavy lines: erosive unconformities, dashed lines: unconformities without evidence of erosion.
horizons used as key beds occur within this unit. The Ichoa Fm. (Cretaceous-Jurassic) is a sandy sequence that contains interbedded shales,
mudstones and gypsum levels. The Carboniferous unit is a continental
mega-sequence made of sandstones with some intercalations of limestones, limolites and gypsum. The Silurian-Devonian mega-sequence
includes five marine, siliciclastic units with limestones concentrated in
the lower part. The Iquiri Fm. (Devonian) consists of two shaly intervals
separated by fine-to very fine-graded sandstones. Since it only appears
in the southernmost well, it was not interpreted on the seismic data and,
therefore, was not considered for the restoration and decompaction.
The Limoncito Fm. (Devonian) is a shaly unit with thin intercalations of
fine-grained sandstones. The Roboré Fm. (Devonian) is mainly formed
by sandstones. The Boomerang Fm. (Devonian) is a shaly level ascribed
to the base of the overlying unit in some wells. The Kirusillas Fm. (Silurian) consists of an alternation of limestones, shales and sandstones
topped by a sandstone bed. The pre-Silurian mega-sequence is made up
of a marine Cambrian-Ordovician succession formed by quartzites with
shale fragments, and Precambrian rocks.
Seismic and well data show that the Mesozoic and Cenozoic megasequences appear everywhere in the study area (Fig. 3). The thickness
of the Cenozoic mega-sequence is around 1.2 km in the northeast part
of the study area (SDG-x1, CTO-x1 and SIG-x1 wells) and more than
3.5 km in the southwest (ICH-x1 well), although its total thickness is
unknown given that the succession top is eroded. The Cenozoic base is a
paraconformity over the Mesozoic. The Mesozoic mega-sequence displays an approximately constant thickness of almost 850 m all over the
study area (Fig. 3), except for some localities where Mesozoic rocks fill
in depressed blocks, in some cases bounded by faults. The base of the
Mesozoic is an erosive angular unconformity that truncates progressively older rocks towards the north from the Carboniferous to the preSilurian mega-sequences (Figs. 3 and 4), and whose angle with the
underlying units is up to 10° increasing to the north. The Carboniferous
and Silurian-Devonian mega-sequences display an irregular distribution
along the studied area. They are almost absent in the north portion of
the study area, where only small remnants of the Silurian-Devonian
mega-sequence may be found in depressed regions controlled by
downthrown fault blocks (Fig. 3). In the south part of the studied region, the Silurian-Devonian and the Carboniferous mega-sequences
have a wedge shape, thickening towards the south and reaching a total
thickness of up to around 8 km. The base of the Carboniferous is an
angular unconformity over the Silurian-Devonian rocks. The top of the
pre-Silurian mega-sequence in the south portion of the study area is an
inclined unconformity where the lower part of the Silurian-Devonian
mega-sequence onlaps, whereas it is roughly flat-lying in the north part
of the study area.
4.2. Mechanical stratigraphy and tectonics-sedimentation relationships
Two units are recognized in the south part of the study area taking
into account how the rocks reacted to deformation: an upper unit
formed by the Cenozoic, Mesozoic, Carboniferous and SilurianDevonian mega-sequences, and a lower unit formed by the pre-Silurian
mega-sequence (Fig. 3). The differentiation of these two units is based
on: 1) the fact that the structures developed in the upper unit are different (mainly folds and faults) and more abundant than those developed in the lower unit (mainly faults), and 2) the occurrence of a significant detachment along Silurian ductile rocks separating the upper
and lower units. The different behavior of these two units is probably
due to a rheological control that caused the upper unit, formed by
shales, sandstones and some limestones and gypsum, to behave in a
relatively ductile manner, while the lower unit, constituted mainly by
quartzites, behaved more rigidly. On the other hand, the high viscosity
contrast between the Silurian shales, located at the base of the upper
unit, and the underlying Ordovician quartzites gave rise to a detachment.
Regarding the tectonics-sedimentation relationships, most of the
Cenozoic rocks (Chaco and Yecua fms.) correspond to a post-extension,
pre-inversion succession. It is debatable whether the youngest sediments of the Chaco Fm. show evidence of syn-tectonic sedimentation
related to contractional/inversion folds, but unfortunately, the shallow
part of the seismic data is poorly imaged. The lowermost Cenozoic sediments (Petaca Fm.) belong to a syn-extension succession (see also
4
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 4. Sub-crop map under the Mesozoic unconformity in the study area (see location of the study area in Fig. 1). The Mesozoic rocks lay on top of progressively
older Paleozoic rocks towards the north. Uncolored regions correspond to areas where no data are available.
detachment in the Boomerang Hills corresponds to the basal detachment of the Andean Cordillera. The orientation of the detachment tip
line varies from E-W to the west of the study area, to ENE-WSW in the
central portion, to E-W towards the east (Figs. 2 and 5). The detachment
tip line follows tightly the zone where the basement top changes from
dipping to the S or SE (south part of the study area) to be sub-horizontal
(north part of the study area). The average strike of the detachment,
and that of the basement top, in the southern region is E-W dipping
gently to the S. The dip of the basal detachment is not constant everywhere; thus, it is gently waved in some areas due to basement irregularities of the order of 3–4 km width, interpreted as old basement
faults and/or Silurian normal faults (Laffitte et al., 1998). The basal
detachment climbs up section near its termination and exhibits a reverse offset in the Mesozoic and Cenozoic mega-sequences (Fig. 6).
However, the cut-off point of the Silurian-Devonian boundary on the
detachment hangingwall is deeper than its footwall equivalent point
denoting that the detachment behaved as a normal fault. The normal
motion probably took place after Silurian-Devonian sedimentation because these rocks do not exhibit syn-extensional features and before
Mesozoic sedimentation, while reverse fault reactivation occurred
during Cenozoic times. Despite the reverse motion along the basal detachment, Devonian and Silurian hangingwall rocks are still sunk with
respect to their footwall equivalent rocks (i.e. part of the normal motion
is preserved) indicating mild inversion tectonics.
The shortening in the frontal part of the basal detachment, i.e. in the
zone where the basement dip changes from dipping gently to the S to
sub-horizontal, is accommodated by two different types of kilometerscale structures, which constitute the northern boundary of the
southern structural domain: 1) thin-skinned tectonic wedges (include
back-vergent structures) (Fig. 6), and 2) thin- and thick-skinned folds
and faults (all the structures verge towards the tectonic transport sense)
(Fig. 7).
Tectonic wedges occur for almost 10 km along strike in the central
portion of the study area, where the tip line of the basal detachment
strikes E-W (Figs. 2 and 5). Two different types of frontal tectonic
wedges occur along strike: 1) Tectonic wedges that accommodate all
the shortening (Fig. 8a), and 2) tectonic wedges where the shortening is
partially accommodated by the wedge and part is transferred to the
north through and upper detachment (Fig. 8b). In the second case, the
basal detachment near its termination becomes a hangingwall ramp
that cuts off the Silurian-Devonian succession developing a fault-bend
fold, and farther north it becomes an upper detachment that runs along
Welsink et al., 1995a). Most of the Cretaceous rocks (Cajones and
Yantata fms.) consist of a pre-extension succession regarding the lowermost Cenozoic and a post-extension succession with respect to the
lowermost Cretaceous and Jurassic rocks (Ichoa Fm.), which are considered as a syn-extension succession (see also Welsink et al., 1995a).
The relationships between tectonics and the Carboniferous and Devonian sediments (Carboniferous, Iquiri, Limoncito, Roboré and Boomerang fms.) are unclear, although they may belong to a pre-extension
succession with respect to the Jurassic-lower Cretaceous succession,
and to a post-extension succession with respect to Silurian sediments
(Kirusillas Fm.) considered as a syn-extension succession (see also
Welsink et al., 1995a). The upper part of the Cambrian-Ordovician
rocks would be a syn-extension succession, whereas the lower part of
the Cambrian-Ordovician and the Precambrian rocks may correspond to
a pre-extension succession.
5. Main structural elements
According to the geological interpretation of the seismic data, the
studied area includes folds as well as normal, reverse, oblique and reactivated faults. Some of these structures affect only the basement,
whereas others are developed in the cover and others involve both
units. Structural maps of the study area (Figs. 2 and 5) show that most
of the main or, at least, larger structures exhibit a general E-W orientation, although some structures strike ESE-WNW in the eastern
sector, whereas they tend to be ENE-WSW in the southwestern sector.
The disparities between the north and south portions of the studied area
led Welsink et al. (1995a) to differentiate two domains, separated by a
major tectonic structure, whose main structural characteristics are described below.
5.1. Structure of the southern domain
The maps in Figs. 2 and 5 show that most of the structures in this
domain are developed either within the cover rocks, or within the cover
and basement, and only a few involve solely the basement.
One of the most prominent structural features of this domain is a
basal detachment (Mingramm et al., 1979; Baby et al., 1989; Dunn
et al., 1995; Kley, 1999) of, at least, 80 km along-strike length, placed
mainly within the shales of the lowermost part of the cover rocks (Silurian), close and subparallel to the boundary with the pre-Silurian
basement. According to Giraudo and Limachi (2001), the basal
5
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 5. Structural map of the Ichoa Fm. in the study area (see location of the study area in Fig. 1) showing the most important folds and faults that affect this
stratigraphic unit. See the stratigraphic position of the Ichoa Fm. in Fig. 3.
the Mesozoic unconformity and places Mesozoic and Cenozoic rocks in
a hangingwall flat situation on top of Silurian or pre-Silurian rocks in a
footwall flat situation. In both types of tectonic wedges, the backthrust,
whose maximum S-directed displacement is approximately 1.3 km,
propagates through the Mesozoic and the lower part of the Cenozoic
succession in a hangingwall ramp over footwall ramp situation causing
an open ramp anticline developed on the backthrust hangingwall,
whose geometry resembles that of shear fault-bend folds (Suppe et al.,
2004). The development of one or another type of tectonic wedge is
controlled by the spatial distribution of the Silurian ductile rocks near
the basal detachment tip. The first mode developed in those areas
where the Silurian ductile rocks pinch out against the south dipping
basement (Fig. 8a). The second mode takes place when the Silurian
ductile rocks exceed the boundary between the S-dipping and subhorizontal basement (Fig. 8b).
In the easternmost and western portions of the study area where the
Fig. 6. Geological interpretation of a seismic line in the central-eastern part of the study area illustrating the boundary between the southern (mainly compressive
and thin-skinned) and the northern (dominantly extensional and thick-skinned) domains. A backthrust emanates from the basal detachment termination forming a
tectonic wedge. A north-dipping normal fault in the northern domain offsets the Pre-Silurian, Silurian, Devonian and Mesozoic mega-sequences and exhibits a thicker
Petaca Fm. in its hangingwall than in its footwall which suggests that fault activity took place during Early Cenozoic. PLC: Palacios well drilling the Palacios anticline
and ECD: Enconada well drilling the Enconada anticline. See location in Fig. 2.
6
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 7. Geological interpretation of a seismic line in the central-western part of the study area showing the boundary between the southern (mainly compressive and
thin-skinned) and the northern (dominantly extensional and thick-skinned) domains. N-directed reverse faults and folds develop near the termination of the basal
detachment. SIR: Sirari well. See location in Fig. 2.
tip line of the basal detachment strikes E-W and ENE-WSW respectively
(Figs. 2 and 5), the frontal termination of the basal detachment forms
two types of fold/fault structures for almost 70 km along strike: 1) folds
in the hangingwall of reverse faults, developed in the vicinity of the
basal detachment tip, that cut and offset the lowermost portion of the
cover and basement rocks located below the basal detachment (Fig. 8c);
and 2) cover rock folds developed near the tip of the basal detachment
(Fig. 8d). In the first case, the reverse faults dip steeply and the fold
backlimbs are oblique to the faults (Fig. 8c). In the second case, the
thrust faults dip moderately and the fold backlimbs are approximately
parallel to the thrust. Both types of folds are N-vergent and open
structures, and both types of faults dip to the south, are N-directed and
their displacements do not exceed 150–200 m.
Frontal tectonic wedges developed only in those areas where the basal
detachment strikes E-W and where the Silurian-Devonian rocks onlap
abruptly on the basement top. In these regions the boundary between the
S-dipping and the sub-horizontal basement top is a narrow, angular zone,
i.e. a kink-like hinge. The absence of favorable ductile Silurian shales,
where the basal detachment is located, inhibited the forward propagation
of the detachment and triggered the development of a backthrust developing a tectonic wedge (Figs. 6, 8a and 8b). In contrast, frontal fold/fault
structures developed where the onlap of the Silurian-Devonian rocks on
the basement top is smooth. In these regions the boundary zone between
the S-dipping and the sub-horizontal basement top is a wide, smooth zone,
i.e. a rounded-like hinge (Figs. 7, 8c and 8d).
Apart from the basal detachment and their related frontal structures, other structures occur within this domain. There are also some EW and ENE-WSW striking faults, usually steeply to moderately dipping
to the S although some N-dipping faults also occur, that may reach
10–15 km along-strike length. These faults are interpreted as old
normal faults affecting the basement. Some of them have been partially
reactivated and propagated up section forming folds, which might be
interpreted as forced folds, affecting all the cover rocks. Their maximum reverse displacement is approximately 200–300 m (basement
faults in Fig. 2).
Not all the structures developed in the southern structural domain
follow the general E-W trend; thus, some faults strike approximately
N–S and dip steeply both to the E and to the W (Fig. 2). These faults
may reach almost 7 km along-strike length, exhibit maximum displacements of approximately 150–200 m and, although they usually
only affect the basement, some of them involve up to the Cenozoic
rocks. Their movement sense is unclear, although some seem to display
a reverse sense of motion.
Short anticlines along strike, whose periclinal terminations are very
close and remind elongated domes, occur in some localities. These relatively symmetrical, open structures, exhibit kilometer-scale dimensions and are well developed in the cover rocks (Figs. 2 and 5). The
internal structure of some of these structures may include steep, reverse
faults with a very small displacement that affect the lowermost part of
the cover (Silurian and Devonian); it is unclear whether they involve
7
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 8. Simplified models simulating the principal terminations of the basal detachment in the Boomerang Hills: a) tectonic wedge where the backthrust emanates
from the basal detachment tip located within a Silurian pinch out, b) tectonic wedge where the backthrust emanates from an upper detachment that terminates above
a half-graben filled in by Silurian rocks, c) folds related to steeply dipping faults that involve basement rocks, and d) folds related to gently dipping faults developed
within cover rocks.
Paleozoic rocks. The last two cases correspond to a mild positive inversion tectonics (usually less than 100 m of fault displacement measured in Cenozoic rocks).
A few E-W trending, S-dipping thrusts, of a few hundred meters
along-strike length, have been also mapped in this domain. Their Ndirected displacements may reach around 50 m and they offset only
cover rocks.
Moreover, a few open to gentle folds, developed in cover rocks, are
mainly located in the south part of this domain. They exhibit subvertical axial surfaces and fold traces that reach a few hundred meters
along strike.
Thus, most of the structures in this domain developed from
Paleozoic to Early Cenozoic (pre-Andean times), except for reactivations of normal faults as reverse faults, thrusts and folds that developed
during Cenozoic (Andean times) because these rocks are the youngest
ones affected by these structures.
From the description of the structures above, we conclude that there
are important differences between the northern and southern structural
domains in terms of structural style, types of structures, tectonic regime
responsible for the structures, age of the structures, basement disposition and involvement in the structures, and stratigraphic units and
thickness (Table 1).
the basement (Fig. 9). These faults dip to the interior of the anticlines
and run mainly along the anticline limbs, and sporadically offset the
anticline crests. Some anticlines display a box-type geometry with a flat
crest in the lowest horizons, whereas they become open rounded folds
in the highest horizons.
Thus, except for some faults that only affect basement rocks, which
could have been active during Paleozoic times (pre-Andean), the age of
the deformation in the southern domain is mainly Upper Cenozoic
(Andean) because these rocks, deposited during the Andean cycle, are
the youngest ones involved in the structures.
5.2. Structure of the northern domain
Unlike in the southern domain, most of the structures in the
northern domain involve both the basement and the cover or only the
basement, and just a few structures affect only cover rocks (Figs. 2 and
5).
The most abundant structures are E-W trending normal faults
steeply to moderately dipping both to the N and to the S. These faults
can reach more than 60 km along-strike length and their maximum
displacement is approximately 300–400 m. The thicker hangingwall
successions with respect to their footwall equivalents (syn-extension
sediments), as well as the age of the faulted rocks, indicate that some
faults were active growth faults during Cambrian-Ordovician times
(structures within the basement), others during the middle part of the
Mesozoic, and others during the early Cenozoic (Fig. 6). Some normal
faults bound remnants of rocks of probably Silurian age, surrounded by
basement rocks but do not offset the erosional unconformity at the base
of the Mesozoic; this suggests that they were active during Paleozoic
times and stopped before the Mesozoic erosion. Some normal faults
exhibit a listric geometry and open anticlines developed in their
hangingwalls, interpreted as rollover anticlines.
Three principal modes of inversion tectonics have been identified,
all of them involving Mesozoic normal faults: 1) Mesozoic normal faults
reactivated as normal faults during early Cenozoic (Fig. 10a), 2) Mesozoic normal faults reactivated as reverse faults during Cenozoic
(Fig. 10b), and 3) Mesozoic normal faults cut and offset by Cenozoic
thrust faults (Fig. 10c). In the first case, the faults show syn-extension
sediments of two different ages (Jurassic-lowermost Cretaceous and
lowermost Cenozoic) separated by a post-extension sequence with respect to the underlying sediments and pre-extension with respect to the
overlying sediments. In the second case, the faults exhibit reverse displacement when involving Cretaceous and Cenozoic rocks, but normal
displacement still preserved in the lowermost Cretaceous-Jurassic and
5.3. Tectonic transport vector
The Andean vector of tectonic transport in the study area and surroundings has been measured using GPS velocities of plate current
movements (Bevis et al., 2001; Hindle et al., 2002), bed-length restorations for different time periods (Hindle et al., 2002), surface restorations
(Hinsch et al., 2003), and measurements of kinematic indicators on fault
populations developed during different times (Mercier et al., 1992; Klosko
et al., 2002). The average direction of the tectonic transport vector estimated using previous authors data is N047E. Since the average direction of
the tectonic transport vector is NE-SW and the strike of the structures
mapped on the seismic data varies from ESE-WNW to ENE-WSW, being EW an average strike, it is likely that the faults did not behave as purely dipslip but oblique faults. Thus, the Boomerang Hills were subjected to
transpression during the Andean orogeny.
Assuming that the average direction of the tectonic transport estimated is correct and constant in the study area, and knowing the orientation of the basal detachment, it is possible to determine the
amount of dip-slip and strike-slip components of the basal detachment
motion during Andean times. These components are controlled by both
the fault dip and the angle between the fault strike and the tectonic
8
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 9. Geological interpretation of a seismic line across the southern domain showing folds whose limbs are cut and offset by faults. ECD: Enconada well drilling the
Enconada anticline and SIR: Sirari well. See location in Fig. 2.
kinematic indicators would be slightly greater than 45°, i.e. the dip-slip
component would be approximately 52% of the total, whereas the
strike-slip component would be of the order of 48%. Steeper (around
60° dip) faults with similar E-W strike show pitches of kinematic indicators greater than 60°, and therefore, they would also be classified as
reverse left-lateral faults, but their percentages of dip-slip component
would be greater, about 70% and, consequently, their strike-slip component would be less than 30% (Fig. 11). The angle between the basal
detachment strike in the western part of the study area, where its strike
is approximately ENE-WSW, and the average tectonic transport vector
is around 15°. This means that the detachment behaved as a left-lateral
reverse fault because the pitch of the kinematic indicators would be
around 15° (Fig. 11); i.e. the strike-slip component would be approximately 82% of the total, whereas the dip-slip component would be
around 18%. This change in the behavior of the detachment along
strike is consistent with the model proposed by Hinsch et al. (2002). It
is likely that several pre-Andean extensional faults also behaved as
oblique faults during the Andean contraction, but in those cases where
the faults maintained their normal displacement despite the Andean
shortening, then the Andean oblique reactivation went unnoticed.
The termination of the basal detachment as a frontal tectonic wedge
has been recognized only in those areas where the basal detachment
strikes E-W. This may suggest that this type of structure only developed
in regions where the basal detachment acted mainly as a reverse fault,
but not in those areas where it had a dominant strike-slip displacement.
Unfortunately, we do not know the orientation of the tectonic
transport vector during pre-Andean times. Since extensional events of
different ages took place, it is possible that the old tectonic transport
vectors had different orientations. In any case, since many pre-Andean
structures exhibit E-W strikes, it is likely the existence of approximately
N–S tectonic transport vectors perpendicular to the structures.
Fig. 10. Simplified models simulating the principal modes of inversion tectonics in the northern domain of the Boomerang Hills. Cretaceous-Jurassic
normal faults: a) reactivated as normal faults during the lower part of the
Cenozoic; b) reactivated as reverse faults during the Cenozoic, so that part of
the normal offset is still preserved, and including backthrusts; and c) cut and
offset by Cenozoic thrusts that dip in opposite sense to the normal faults.
transport vector (Fig. 11). In general, the percentage of dip-slip component increases as the fault dip and the angle between the fault strike
and the tectonic transport vector rise. The true strike and dip of the
basal detachment were estimated in two localities using depth-converted seismic data. The angle between the average tectonic transport
vector and the strike of the detachment in the eastern part of the studied area, where it strikes approximately E-W, is around 45° and the
detachment is subhorizontal. According to the graph (Fig. 11), the detachment behaved as a reverse left-lateral fault because the pitch of the
6. Construction and restoration of a 2D cross-section
Since the geological interpretation was done in time, we needed a
velocity model to depth-convert the interpretation and the compaction
parameters to analyze the thickness variations in the stratigraphic sequences through time.
9
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Table 1
Main differences between the southern and northern structural domains of the Boomerang Hills.
Main structural style
Most common type of structures
Detachment features
Dominant tectonic regime
Age of main deformation
Main orogenic cycle
Most important pre-Andean structures
Most important Andean structures
Basement top attitude
Geometry of the cover units
Units more frequently involved in the structures
Age of the stratigraphic sequence
Thickness of the stratigraphic sequence
Tectonic unit
Most common hydrocarbon
SOUTHERN DOMAIN
NORTHERN DOMAIN
Mainly thin-skinned
Folds and faults
S-dipping, detachment
Contractional
Cenozoic
Andean
Paleozoic extensional detachment
Frontal detachment termination (tectonic wedges and faults/
folds)
S-dipping
Wedge shaped (lower part) and tabular (upper part)
Cover and basement-cover
Basement: Precambrian-Ordovician
Cover: Silurian, Devonian, Carboniferous, Mesozoic, Cenozoic
Thick
Andes
Gas
Thick-skinned
Faults
No detachment
Extensional
Paleozoic-Early Cenozoic
Pre-Andean
Normal faults
Reverse reactivation of normal faults, and minor thrusts and
folds
Sub-horizontal
Tabular
Basement and basement-cover
Basement: Precambrian-Ordovician
Cover: Silurian (sporadically), Mesozoic, Cenozoic
Thin
Beni-Chaco Plain
Oil
stratigraphic tops and/or faults, had already been mapped. The depth
conversion was carried out using the software Move©. Anyway, we
must bear in mind that the depth conversion performed is an approximation.
6.2. Compaction parameters
When performing a restoration the stratigraphic units must undergo
a certain amount of decompaction as the ones overlying them are removed. The variation of porosity with depth for each stratigraphic unit
must be known in order to take compaction into account. This
information is usually calculated by fitting the following exponential
function through points plotted on porosity versus depth graphs
(Eqn 1):
=
0
e
(1)
cy
where φ is the porosity, φ0 is the initial porosity, c is the compaction
coefficient, and y is the depth. Seventeen well reports include tables
gathering values of porosity and depth for different Cenozoic, Mesozoic,
Devonian and Silurian units. The main drawback that arose when trying
to find out the exponential functions for each stratigraphic unit was that
most of the well data were constrained to narrow depth ranges and did
not cover a sufficient depth range to support any particular function
(full circles in Fig. 12a). This made difficult the determination of the
initial porosity and the compaction coefficient. To sort this out, on
depth versus porosity graphs we freehand drew a function (solid line in
Fig. 12a) through the average value of the well-data point cloud following a similar path to that of the functions constructed by previous
authors for lithologies equivalent to those of the stratigraphic unit to be
analyzed (dashed lines in Fig. 12a). To quantify the initial porosity and
the compaction coefficient, points at arbitrary, different depths on the
Fig. 11. Graph that illustrates the pitch of the kinematic indicators on the fault
surface versus the angle between the fault strike and the tectonic transport
vector for different fault dips. The values corresponding to three structure types
found in the Boomerang Hills are plotted on the graph. To construct this graph,
the orientation of the line resulting from the projection of the tectonic transport
direction on the fault plane has been calculated (the direction of this line is that
of the transport direction and its dip is the apparent dip of the fault plane along
that direction), then the angle between the fault strike and this line on the fault
plane (i.e. the pitch) has been calculated for different fault dips.
6.1. Velocity model and depth conversion
The horizons and surfaces obtained from the geological interpretation of the seismic data were depth converted using the information
from ten wells in order to visualize the proper geometry of the structures in both 2D and 3D, as well as to carry out the restoration of a
geological cross-section and a 3D geological model. Three types of velocity models were constructed using these values: velocity laws for
each well, an average velocity law combining the laws derived from all
the wells, and interval velocities for each seismic-stratigraphic unit
obtained by averaging the velocities of each unit estimated in the
available wells. The models based on velocity laws were not employed
because the large variations in the stratigraphic thicknesses of some
units throughout the study area would invalidate their use. Therefore,
the velocity model, based on interval velocities (Table 2), was selected
to carry out the depth conversion, as it should offer the best results
given that the boundaries of the different velocity units, i.e.
Table 2
Interval velocities calculated for each stratigraphic unit interpreted on
the seismic data.
10
Unit
Velocity (m/s)
Chaco Fm. (Cenozoic)
Yecua Fm. (Cenozoic)
Petaca Fm. (Cenozoic)
Yantata Fm. (Cretaceous)
Ichoa Fm. (Cretaceous-Jurassic)
Carboniferous
LImoncito Fm. (Devonian)
Roboré Fm. (Devonian)
Boomerang Fm. (Devonian)
Kirusillas Fm. (Silurian)
2531
3129
3263
3227
3446
4161
3649
4023
3582
4034
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
structures and stratigraphic features of the studied area from the
southern part, where Andean structures dominate, to the northern part,
where pre-Andean structures are the main ones, and intersects or passes
near a number of wells (Fig. 2). Since the orientation of the geological
cross-section cannot be both parallel to the NE-SW average direction of
the tectonic transport vector (so that the section is retrodeformable)
and perpendicular to the E-W to ESE-WNW average strike of the
structures (so that the section illustrates their proper geometry), the
plane selected to build the cross section is vertical and its strike is
N030E. The cross section was constructed by projecting the geological
interpretations of the closest seismic profiles onto the selected crosssection plane, because none of the seismic profiles has the orientation of
the chosen geological section.
6.4. Cross section description
The main structures in the central part of the cross section (Fig. 13)
are: 1) two gentle to open detachment folds related to motion above the
basal detachment located close to the base of the Kirusillas Fm. (Silurian), just above the basement top; and 2) a tectonic wedge including
a possible shear-fault bend fold related to a backthrust which emanates
from the basal detachment tip where the Paleozoic succession no farther north occurs. The most relevant structures in the northeast portion
of the cross section (Fig. 13) are steep normal faults that cut and offset
the uppermost part of the basement, the Mesozoic units and the lowermost Cenozoic unit (Petaca Fm.). These faults may have been active
during the deposition of the Ichoa Fm. (Jurassic-Cretaceous) and the
Petaca Fm. (Cenozoic), given that these stratigraphic units are thicker
in their fault hangingwalls.
In addition to the variation of the structural style from southwest to
northeast, there is also an important stratigraphic change (Fig. 13). The
base of the Mesozoic rocks is an erosional unconformity, so that Mesozoic units lay directly on top of the pre-Silurian basement to the
northeast, where Silurian-Devonian and Carboniferous rocks are absent,
whereas they are deposited above progressively younger Paleozoic
rocks to the southwest from the Kirusillas Fm. (Silurian) in the central
part of the cross section to the Carboniferous in the southwest part of
the section (Fig. 4). Furthermore, the Paleozoic sequence exhibits a
wedge shape, so that it is thicker and more complete to the southwest.
Fig. 12. a) Depth versus porosity graph for the Petaca Fm. Full circles: well
data, dashed lines: reference curves taken from the literature (Bond et al., 1983;
Angevine et al., 1990) for lithologies similar to those of the Petaca Fm., solid
line: extrapolated curve, hollow circles: new points to be fitted by an exponential curve in b). b) Porosity versus depth graph including an exponential
function that fits the new points for the Petaca Fm displayed in a). See the
stratigraphic position of the Petaca Fm. in Fig. 3.
Table 3
Initial porosity (φ0) and compaction coefficient (c) estimated for the different
stratigraphic units using the methodology described in the text.
Unit
φ0
c
Chaco Fm. (Cenozoic)
Yecua Fm. (Cenozoic)
Petaca Fm. (Cenozoic)
Yantata Fm. (Cretaceous)
Ichoa Fm. (Cretaceous-Jurassic)
Carboniferous
LImoncito Fm. (Devonian)
Roboré Fm. (Devonian)
Boomerang Fm. (Devonian)
Kirusillas Fm. (Silurian)
0.38
0.38
0.38
0.40
0.39
0.36
0.36
0.39
0.36
0.36
0.331
0.331
0.331
0.299
0.259
0.371
0.371
0.383
0.371
0.371
freehand drawn function (hollow circles in Fig. 12a) were plotted on a
porosity versus depth graph and fitted by an exponential best-fit function (Fig. 12b). When no well data were available for a particular
stratigraphic unit, well values obtained for lithologies comparable to
this particular stratigraphic unit were employed. The initial porosity
and the compaction coefficient for each stratigraphic unit derived from
the method described above are displayed in Table 3.
6.5. Sequential restoration and decompaction of the cross section
The depth-converted 2D geological transect across the study area
(Fig. 13), was sequentially restored to show the evolution of this portion of the study area. The restorations were performed using the
geometric/kinematic algorithms called “flexural-slip” and “fault-propagation fold” implemented in the software Move©, and using a mechanical methodology based on the finite element method (FEM) applied to linear elastic deformation implemented in the software Dynel©
(Maerten and Maerten, 2001). These 2D restoration algorithms were
6.3. Selection of the cross section location and orientation
The selected cross section runs across the most characteristic
Fig. 13. Geological cross-section constructed by projecting the geological interpretations of nearby seismic profiles onto a N030E vertical plane. PLC: Palacios well
drilling the Palacios anticline and LPS: Los Penocos well. See location in Fig. 2.
11
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Table 4
Mean “true” (i.e. normal to the stratigraphic unit top) thickness in meters of the different stratigraphic units nowadays (0) and after each decompaction step (1–9).
Unit
0
1
2
3
4
5
6
7
8
9
Chaco Fm.
Yecua Fm.
Petaca Fm.
Yantata Fm.
Ichoa Fm.
Carboniferous
Limoncito Fm.
Roboré Fm.
Boomerang Fm.
Kirusillas Fm.
2463
554
416
320
511
1023
547
887
336
1255
–
757
520
387
574
1131
578
951
353
1315
–
–
590
434
667
1191
598
980
362
1340
–
–
–
476
668
1230
617
1004
370
1359
–
–
–
–
721
1306
644
1032
381
1381
–
–
–
–
–
1475
743
1112
410
1429
–
–
–
–
–
–
976
1202
429
1478
–
–
–
–
–
–
–
1332
461
1535
–
–
–
–
–
–
–
–
648
1695
–
–
–
–
–
–
–
–
–
1794
used because they are the ones that best emulate the particle motion
caused by the most important structures in the geological cross-section.
The pin line was placed at the northeast end of the geological crosssection and the loose line at the southwest edge. To show the thickness
variations over time of the stratigraphic units, they were decompacted
during the restoration using the software Move© and employing the
functions obtained by ourselves for each stratigraphic unit (Table 4).
The first restoration step consisted of removing the displacement
along the detachment and the backthrust and unfolding both the shear
fault-bend fold and the detachment folds in the central part of the cross
section; these must be the youngest structures since they involve Upper
Cenozoic rocks (Fig. 14a). The mechanical restoration of these structures shows that strain was especially intense at the backlimb of the
shear-fault bend fold and in the backthrust footwall close to the point
where it merges with the basal detachment (Fig. 14b). The absolute
value of the maximum strain achieved is 0.50. The strain computed in
the mechanical restoration is the deformation suffered by the linear
triangular elements in which the geological cross-section has been
discretized in order to build a mesh (Maerten, 2010). The shortening
values measured as the length of the restored section minus the length
of the present-day, deformed section range from 1.5 km, in the case of
the geometric/kinematic restoration, to 1.0 km, in the case of the mechanical restoration. Different values of shortening are obtained because all the shortening is consumed in displacement of the rocks to
form thrusts and folds in the geometric/kinematic restoration, whereas
part of the shortening is consumed in displacement to form thrusts and
folds, and the remaining shortening is accommodated as strain suffered
by the rocks in the mechanical restoration. The numerical results obtained from the restorations are not totally accurate due to possible
errors made during the geological interpretation of the seismic data, the
projection of the geological interpretations onto the chosen plane, the
depth-conversion, the estimation of the decompaction functions and the
angle of 17° between the section line and the average direction of the
contractional tectonic transport vector. After returning the Cenozoic
units to its initial horizontal position, these stratigraphic units were
removed and the underlying succession was decompacted accordingly.
The displacement along the normal faults located in the northeast
part of the cross section was removed in the next restoration step. This
was performed in two stages, as they seem to have been active during
two different times: during the deposition of the Petaca Fm. (lower
Cenozoic) and during the sedimentation of the Ichoa Fm. (Jurassiclower Cretaceous) (Fig. 14a). According to the mechanical restoration,
the strain is mainly concentrated in the fault hangingwalls close to the
faults, being 0.16 the maximum absolute value (Fig. 14b). The extension values measured as the length of the restored section minus the
length of the section restored to a younger stage are negligible in both
the geometric/kinematic and the mechanical restorations given that
they are less than 0.1 km. Again, possible errors made during the
geological interpretation of the seismic data, the projection of the
geological interpretations onto the chosen plane, the depth-conversion,
the estimation of the decompaction functions and the 30° angle
between the section line and the suspected direction of the extensional
tectonic transport vector, make the quantitative results derived from
the restorations approximate. As in the previous step, after each restoration the corresponding stratigraphic units were removed and the
underlying units were decompacted.
Once all the folds and faults were restored, the main unconformity,
located at the base of the Mesozoic rocks, was rotated to a horizontal
disposition. After this, the Paleozoic units were sequentially restored to
a horizontal attitude (Fig. 14a), which allowed illustrating the progressive tilting of the basement top in the southwest part of the study
area as the deposition of the Silurian, Devonian and Carboniferous took
place (Fig. 15). The basement top dip varied through time from around
3° to the SW in Ordovician times to about 11° nowadays. According to
the mechanical restoration, the maximum strain took place in the hinge
zone of the gentle flexure that involves the basement top and reached a
maximum value of 0.60 (Fig. 14b). These dip and strain values are
approximate due to the reasons explained above. Since some portions of
the Paleozoic units below the pre-Mesozoic unconformity were eroded,
they were reconstructed before their decompaction to obtain a more
realistic result. The sequential restoration of the Paleozoic rocks allowed the detection of subtle sedimentary geometries similar to progressive unconformities that, otherwise, could have been neglected. For
instance, the Boomerang and the Limoncito fms. (Devonian) are approximately tabular, whereas Kirusillas and Roboré fms. (Silurian and
Devonian respectively) are wedge-shaped (Fig. 13). This thickness
variation pattern may reflect subsidence changes, i.e. periods of relative
quiescence would result in constant thicknesses and almost no rotation
of the basement top, whereas accelerated subsidence might favor
thickening towards the basin depocenter and rotation of the basement
top (Fig. 15).
Thus, according to the restorations carried out, the maximum horizontal displacement values took place as shortening during the Late
Cenozoic compressive episode (Andean), whereas the extension values
during the Early Cenozoic and the Mesozoic extensional episodes (preAndean) were inferior. In the same way, the maximum strain values
occurred during the Andean episode and during the basement tilting
occurred in Paleozoic times.
7. Construction and restoration of a 3D model
7.1. 3D model generation
The geological interpretation of faults carried out on the seismic
profiles and on the 3D seismic volumes was interpolated to create fault
surfaces using the module called convergent interpolation implemented
in the software Petrel©. Once the fault framework was built, the procedure was replicated with each of the tops of the stratigraphic units
and unconformities, taking into account how these surfaces were cut
and offset by the fault surfaces previously generated (Fig. 16). The resulting surfaces were depth-converted using the interval velocity model
presented above to allow the correct visualization of their geometry, to
12
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 14. 2D sequential restoration and decompaction of the geological cross-section shown in Fig. 13 using: a) the software Move© combining the flexural-slip and
fault-propagation fold algorithms, and b) the software Dynel© using a finite element method (FEM) applied to linear elastic deformation. The dashed boundaries of
the Carboniferous, Limoncito Fm. and Roboré Fm. in the north part of the restorations indicate that we ignore whether these stratigraphic units were or not deposited
in this region.
13
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 15. Evolution through time of the mean dip of
the basement top in the south part of the study area
derived from the sequential restoration of the geological cross-section illustrated in Fig. 14 (black solid
line and black triangles: geometric/kinematic restoration, black dashed line and black squares: mechanical restoration), and from the sequential restoration of the 3D model illustrated in Fig. 17 (black
dotted line and black circles: kinematic restoration).
The grey arrows correspond to average tilting paths
followed by the basement top with their corresponding rotation rates.
Fig. 16. 3D geological model of the study area showing the main faults and stratigraphic tops, as well as wells. Some tops, such as those corresponding to intra
Devonian units, have not been shown in the model for the sake of clarity. Faults are displayed in white. The names of the wells are listed in Fig. 2 caption.
be able to measure dips and strikes, and to calculate burial depth and
unit thicknesses. Finally, both the fault and stratigraphic surfaces were
smoothed to avoid drawbacks during the restoration.
and is located to the north of the study area where almost no structures
occur. Given that the second restoration stage consisted mainly of untilting progressively the basement top by removing the Silurian, Devonian and Carboniferous, wedge-shaped units (Fig. 17), the pin surface
chosen is vertical and follows the hinge of the basement flexure (i.e. the
transition from S-dipping to sub-horizontal basement top). The unfolding plane chosen (060/90) bisects the main bend of the hinge trace
of the basement flexure in the central part of the study area (Fig. 2). The
normal faults that offset part of the Cenozoic and Mesozoic sequences
were not considered in this restoration given their small displacements.
Moreover, the stratigraphic units were decompacted after each restoration step using the software Move© and the functions estimated
above to determine their thickness (both vertical and normal to the top)
at each time interval (Table 4).
The variation of the basement top dip in the south part of the study
area was plotted on the graph in Fig. 15 to illustrate how this dip increased as the Paleozoic units were deposited. The average dip of the
basement top varied from around 6° to the SW in Ordovician times to
7.2. Sequential restoration and decompaction of the 3D model
Similarly to the cross section restoration, the depth-converted 3D
model, i.e. the present-day, deformed 3D model (Fig. 16), was sequentially restored employing a geometric and geometric/kinematic
algorithm called “flexural-slip” implemented in the software Move©.
This 3D restoration algorithm was selected because it adequately simulates the activity of the main structures represented in the 3D geological model. Different pin and unfolding planes/surfaces were used
for each of the main restoration steps. To restore the Cenozoic structures (first restoration stage), i.e. unfolding the cover folds related to
motion along the basal detachment, the unfolding plane has the same
orientation than the plane of the restored 2D cross-section, which is
120/90. The pin plane (030/90) is perpendicular to the unfolding plane
14
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 17. 3D sequential restoration of the basement top surface extracted from the 3D model depicted in Fig. 16, carried out using the flexural-slip algorithm
implemented in the software Move©. Contours each 250 m and vertical exaggeration x2.
about 14° nowadays. These values, derived from the 3D restoration, are
slightly higher than those obtained from the 2D restoration. The reason
is that the restored geological cross-section is oblique to the line of
maximum dip of the basement top, and therefore, it shows apparent
dips, whereas the dips extracted from the 3D geological model are true
dips. Despite being true dips, these values may not be totally correct
due to possible errors made during the geological interpretation of the
seismic data, the construction of the 3D geological model, the depthconversion, the estimation of the decompaction functions and the
choice of the pin and unfolding planes/surfaces.
The graph in Fig. 15 shows various stages with notably different
rates. The basement top dip would have increased at an average rate of
approximately 0.08°/My during Silurian and Devonian, whereas it
would have increased at a slower average rate of around 0.02°/My
during Carboniferous and, probably, Permian and Triassic. The higher
rotation rate of the basement top during Silurian and Devonian is interpreted as a result of rapid basin subsidence, while the lower rotation
rate during Carboniferous and, probably, Permian and Triassic suggests
slower basin subsidence. We are not sure about the subsidence causes,
but the sedimentary load may have been one of them, given that the
average sedimentation rate was 48 m/My during Silurian and Devonian, while it was 25 m/My during Carboniferous according to the
decompacted thickness in Table 4. Another hypothesis to explain the
rotation rate variations might be a widespread sedimentation during
Carboniferous and Permian, which, in turn, would produce a generalized subsidence of the area, less focused on the south where the basement top dips to the S. On the contrary, the Devonian and Silurian
sedimentation would have been concentrated in the south part of the
study area (as shown by the Silurian onlaps to the north) generating a
heavier sedimentary load in the southern region.
The basement top dip is slightly higher during the Paleogene than
during the Mesozoic and is slightly lower during the Neogene. The dip
increase during Early Cenozoic might be interpreted as a result of lithosphere bending due to the load exerted by the weight of the sediments and Andean thrust sheets. The dip decrease during Late Cenozoic
might result from to the load reduction due to the exhumation triggered
by the erosion of the relief created during the Andean orogeny.
7.3. Variation of unit thickness and burial depth through time
Table 4 shows the variation of the mean thickness for each stratigraphic unit as the overlying units were removed and the remaining
units were decompacted using the decompaction functions estimated
above. Two thickness values were computed: vertical thickness and true
thickness (normal to the stratigraphic unit top). Only the true thickness
variation is presented since the differences between the true and the
vertical thickness are almost negligible, which is consistent with the
fact that most of the horizons are sub-horizontal or display gentle dips.
The thickness reduction suffered by the stratigraphic units from their
deposition to nowadays is notable since it ranges from 30 to 50%
(Table 5). Maps to illustrate the thickness of each stratigraphic unit
throughout the study area at different times have been constructed by
combining the 3D restoration together with the decompaction suffered
by the units as the overlying units were removed. As an example, the
thickness variation of the Roboré Fm. is depicted in Fig. 18. The maps
show that the thickness variation was more significant in the central
15
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
and southern parts of the study area where the initial thickness of this
formation was greater. Since thickness variation due to decompaction is
related to porosity lost, these maps can help to estimate the amount of
porosity lost of a source or reservoir rock in different areas at each time
interval.
Employing the restorations and considering the decompaction, we
also constructed maps of burial depth over time for the tops of the
different stratigraphic units, i.e. the vertical distance from one particular stratigraphic horizon to the paleotopographic surface. Only the
burial depth maps corresponding to the Boomerang Fm. top are shown
as an example (Fig. 19). The maps illustrate how the wedge shape of the
Silurian, Devonian and Carboniferous rocks caused a depth increment
more important to the south, and how the regular distribution of the
Mesozoic and Cenozoic mega-sequences produced a diffuse increase in
the burial depth. These maps, combined with additional information
such as paleogeothermal gradient and so on, are useful tools to assess
Table 5
Mean accumulated thickness change (%) after deposition of each stratigraphic
unit from the Kirusillas Fm. of Silurian age (9) to the Chaco Fm. of Cenozoic age
(0).
Unit
0
1
2
3
4
5
6
7
8
9
Chaco Fm.
Yecua Fm.
Petaca Fm.
Yantata Fm.
Ichoa Fm.
Carboniferous
Limoncito Fm.
Roboré Fm.
Boomerang Fm.
Kirusillas Fm.
0
−27
−29
−33
−29
−31
−44
−33
−48
−30
–
0
−12
−19
−20
−23
−41
−29
−46
−27
–
–
0
−9
−7
−19
−39
−26
−44
−25
–
–
–
0
−7
−17
−37
−25
−43
−24
–
–
–
–
0
−11
−34
−23
−41
−23
–
–
–
–
–
0
−24
−17
−37
−20
–
–
–
–
–
–
0
−10
−34
−18
–
–
–
–
–
–
–
0
−29
−14
–
–
–
–
–
–
–
–
0
−6
–
–
–
–
–
–
–
–
–
0
Fig. 18. Variation through time of the thickness of the Roboré Fm. of Devonian age (reservoir rock) in the study area (see location of the study area in Fig. 1). See the
stratigraphic position of the Roboré Fm. in Fig. 3.
16
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 19. Variation through time of the burial depth for the Boomerang Fm. top of Devonian age (source rock) in the study area (see location of the study area in
Fig. 1). See the stratigraphic position of the Boomerang Fm. in Fig. 3.
which areas of the source rocks could enter the hydrocarbon maturation
window at each time stage.
and Carboniferous rocks, and the onlap of the Silurian and Devonian on
the basement top where it dips to the south (Figs. 3 and 6–8 and 13 and
16), point out that the basement top acquired its S dip during the Paleozoic, as shown in the restorations (Figs. 14 and 17) and in the graph
in Fig. 15. Therefore, the region where the basement top dips to the S
was the edge of a high in the Silurian-Carboniferous basin. In addition,
this edge behaved as an extensional detachment during Paleozoic times
and was reactivated as a contractional detachment during the Cenozoic
(Fig. 6). Thus, the location, geometry, size and orientation of the frontal
termination of the Andes (Figs. 2 and 5) was basically controlled by the
edge of the Silurian-Carboniferous high (Hinsch et al., 2002), the extensional detachment and, probably the thinning towards the north of
the Paleozoic sequence (Giraudo and Limachi, 2001).
Tectonic wedges formed at the tip of the basal detachment where
the Silurian and Devonian rocks onlap abruptly on the basement top,
whereas faults and folds formed where they onlap smoothly on the
basement top (Figs. 6–8). The tectonic wedge backthrust emanates from
8. Controls on the Andean structure
The basal detachment runs along the shales located in the lower part
of the Silurian succession, i.e. it took advantage of favorable ductile
rocks. It terminates where these rocks are absent or they are very thin
(Figs. 6–8). Thus, the location of the Andean basal detachment within
the stratigraphic succession was mainly controlled by the mechanical
stratigraphy of the Paleozoic succession (Baby et al., 1994; Welsink
et al., 1995a; Kley, 1996; Anderson et al., 2017).
The present-day geometry of the basement top, sub-horizontal to
the north and S-dipping in the southern region, would be the result of
combining a paleorelief inherited from the Brazilian Shield (Hinsch
et al., 2002) with progressive lithosphere flexure due to sedimentary,
and perhaps tectonic, load. The wedge shape of the Silurian, Devonian
17
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
the basal detachment tip when the ductile rocks lay only above the Sdipping portion of the basement top, whereas the backthrust emanates
from an upper detachment connected to the basal detachment through
a ramp when the ductile rocks lay above the S-dipping and above the
sub-horizontal portions of the basement top. Thus, the type of Andean
frontal structures was controlled by the geometry and position of the
Silurian-Carboniferous rocks (Welsink et al., 1995a).
Although new Andean structures occur in the study area, various
contractional structures formed in Cenozoic times resulted from mild,
selective inversion tectonics of extensional structures inherited from
Mesozoic and/or Paleozoic times, either in the form of reverse reactivation or thrusts that cut and offset previous extensional faults
(Fig. 10). This situation is common in the Andean Cordillera, and it has
been documented both at small and large-scale in the area (Sempere
et al., 1991; Welsink et al., 1995a, 1995b; Mescua and Giambiagi, 2012;
Perez et al., 2016). Thus, not only the Andean basal detachment and its
associated frontal structures undergo control by pre-Andean structures,
but other Andean-age structures are also influenced by previous structures.
The arguments above indicate that the location, morphology, dimensions and orientation of the frontal part of the Andes in the study
area depended strongly on the basement configuration as well as on the
Paleozoic stratigraphy. Therefore, the main orientation of the structures
in the Boomerang Hills would not respond to a change in the tectonic
transport direction, but to the control imposed by previous elements.
decrease in the extension intensity and a sea level rise would result in a
less constrained subsidence (Sempere, 1994, 1995).
The basin emerged between Cretaceous and Cenozoic times, and the
depositional environment change caused a paraconformity on which
lower Cenozoic continental rocks sedimented throughout the entire
area over Mesozoic rocks (Fig. 20). The dip in the southern part of the
Boomerang Hills increased slightly during Early Cenozoic and synchronously a few normal faults developed in the northern part, some of
them resulting from reactivation of old Mesozoic normal faults (see also
Welsink et al., 1995a). The dip variation might result from lithosphere
bending because of the sedimentary and tectonic loads caused by the
weight of the Andean syn-contractional sediments and thrust sheets
emplaced in more internal parts of the Cordillera, whereas the normal
faults would have formed in the outer arc of the flexure. During late
Cenozoic times a thick continental sequence deposited under relatively
calm tectonic conditions. The subsequent Andean event, which would
have started around Oligocene-late Miocene in the Central Andes (e.g.
Sempere et al., 1990; Uba et al., 2006) and had propagated to the
Subandean area about 10–12.5 Ma (e.g. Gubbels et al., 1993; Moretti
et al., 1996; Uba et al., 2009; Lease et al., 2016), consisted of oblique
reverse reactivation of the gravity-driven, extensional detachment of
Paleozoic age, which became the Andean basal detachment, as well as
development of folds and thrusts (Fig. 20). The Andean basal detachment became the Andean frontal thrust when propagated up to the
hinge of the flexure that separates the southern slope of the Paleozoic
basin high from the northern, elevated area. Thus, the Andean contractional structures affected mainly the Silurian-Devonian, Carboniferous, Mesozoic and Cenozoic rocks located above the southern slope
of the Paleozoic basin high, although a few new thrusts and selective
reactivation of some old normal faults as reverse faults occurred further
north of the Andean frontal thrust in the portion of the Beni-Chaco
Plain adjacent to the Andes. A slight dip decrease in the southern part of
the Boomerang Hills occurred during Late Cenozoic. As in most
mountain ranges, sedimentation, development of tectonic structures
and denudation of the growing relief would have taken place simultaneously during this period. The dip decrease might be due to the fact
that the balance between sedimentary/tectonic load with respect to
denudation was favorable to the latter.
9. Tectono-stratigraphic evolution
After Precambrian times, an extensional regime evidenced by faults,
occurred synchronously to Cambrian and Ordovician marine sedimentation (Fig. 20). In Ordovician times, a more uniform and widespread sedimentation took place which ended with a notable, regional
angular unconformity during the Late Ordovician due to the Ocloyic
orogeny (Tankard et al., 1995). This situation was replaced by a more
widespread marine sedimentation in late Silurian to Devonian times,
interpreted as the result of a relaxation in the extensional stresses (Bally
and Snelson, 1980). The unconformity in the south part of the Boomerang Hills rotated, possibly due to subsidence, while in the north part
remained approximately stable. The tilted region became the southern
slope of a high, so that marine Silurian and Devonian sediments onlapped on it and thickened towards the basin depocenter located to the
south. Above the high, Silurian rocks filled in small, depressed regions
bounded by faults probably initiated during Silurian times. The contrast
between rapid tilting rates during Silurian and Devonian times, and
slower ones during Carboniferous, and maybe Permian-Triassic, caused
a slight angular unconformity at the base of the continental Carboniferous rocks. Once the edge of the high was relatively steep, i.e. after
sedimentation of Silurian-Devonian and before sedimentation of Jurassic, an extensional detachment, probably due to gravity, developed
along the southern slope of the high (Fig. 20).
A noticeable erosion occurred between Carboniferous and Jurassic
times, possibly resulting from basin uplift, gave place to a marked angular unconformity (see also Tankard et al., 1995) (Fig. 20). Thus, part
of the Precambrian-Ordovician, Silurian, Devonian and Carboniferous
rocks were eroded as well as any Permian and Triassic successions that
had deposited in the region. This process could be related to a reorganization of the stresses in the area during Triassic times (Sempere,
1995). Uplift due to the rifting associated with the Atlantic Ocean
opening in Mesozoic times (Nürnberg and Müller, 1991; Hall et al.,
2018) is one of the hypotheses proposed to explain the erosion. Continental to shallow marine Jurassic and Cretaceous rocks sedimented
unconformably over the underlying Precambrian-Paleozoic succession
irrespective of whether this succession was tilted or untilted. Mesozoic
was a tectonic quiescence period, although new normal faults developed during Jurassic-Early Cretaceous (Fig. 20). Part of the sedimentation was controlled by synsedimentary faults, although a
10. Hydrocarbon potential
The main source rocks for hydrocarbons in the study area are of
Silurian, Devonian and, less important, Carboniferous age (Fig. 3).
Reservoirs have been identified in the Silurian, Devonian (mainly Roboré Fm.), Mesozoic (Yantata and Ichoa fms.) and Cenozoic units
(Yecua and Petaca fms.). Silurian, Devonian, Carboniferous and Cenozoic shales act as seals (e.g. Illich et al., 1981; Baby et al., 1995;
Welsink et al., 1995a; Lindquist, 1998).
Those wells whose available information includes their geographical coordinates and hydrocarbon contents have been plotted on
the structural map of the study area depicted in Fig. 2. The purpose of
this map is trying to understand some relationships between the underlying geological structures and the wells position and hydrocarbon
contents.
Most wells are located in the southern domain of the Boomerang
Hills, i.e. in the Andes front, close to the basal detachment tip where: 1)
the stratigraphic succession is quite complete and, as a consequence,
source, reservoir and seal rocks are well developed (Fig. 3); and 2) there
is a large number of structures, such as folds and thrusts, and therefore,
large probabilities of structural traps (Figs. 2 and 5). Most of the successful wells are aligned following crests of anticlines along strike,
which run approximately parallel to the trace of the basal detachment
tip line, i.e. from E-W to ENE-WSW, and involve the cover mega-sequences, i.e. Silurian-Devonian, Carboniferous, Mesozoic and Cenozoic
(Figs. 2, 6, 7, 9 and 13). In contrast, unsuccessful wells are usually
located away from the crest of the anticlines. There are fewer wells in
18
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Fig. 20. Stratigraphic column of the Boomerang Hills including the large-scale tectonic events, the tectonic regimes, and the structures formed during each event.
Wavy lines: erosive unconformities, dashed lines: unconformities without evidence of erosion. Stratigraphic and time columns on the left modified from Baby et al.
(1995) and Welsink et al. (1995a). Geodynamic setting column on the right modified from Tankard et al. (1995).
the northern domain of the Boomerang Hills, i.e. in the Beni-Chaco
Plain, where: 1) the stratigraphic succession is more incomplete because Paleozoic source rocks are almost missing, except for Silurian
shales preserved in sporadic depressions (Fig. 3); and 2) the density of
tectonic structures, and therefore, of possible structural traps, is lower
than in the southern domain (Figs. 2 and 5). Most of the wells are located along the eastern portion of an ESE-WNW to E-W striking, Ndipping, long normal fault which offsets the pre-Silurian basement, as
well as the Mesozoic and the lower part of the Cenozoic cover megasequences (Fig. 2). These wells drill the fault hangingwall near the fault
surface (Figs. 6 and 13). The central and western parts of this fault have
not yet been explored. Other wells located, presumably, at the crest of
anticlines related to inverted or reverse faults have not found hydrocarbon shows.
Oil discoveries are more frequent in the wells drilled in the northern
domain of the Boomerang Hills, whereas gas is more common in the
19
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
southern domain wells (Fig. 2 and Table 1). Moreover, in the northern
domain all the productive wells contain a certain amount of oil,
whereas in the southern domain there are wells that contain only gas.
Thus, the spatial distribution of gas and oil might be conditioned to a
certain extent by the age of the Andean structures; the oldest structures
would contain both, while the youngest ones may contain gas solely.
From a subsurface study and its comparison with laboratory physical
experiments, Baby et al. (1994) concluded that the hydrocarbon migration in this region ended in the time interval between the development of the Andean contractional structures located in the outer part of
the southern domain and that of the innermost ones, that is, in Late
Cenozoic. The fact that hydrocarbons are found in the hangingwall of
normal faults developed during Early Cenozoic suggests that, if only a
single hydrocarbon migration stage took place, it began by the Early
Cenozoic and ended in the Late Cenozoic.
the decompaction, has proven to be a valuable tool in order to characterize the main structures, as well as decipher the structural history of
the Boomerang Hills. The results obtained may be a first step to better
understand the hydrocarbon distribution in this province, and therefore, to improve the exploitation of discovered hydrocarbon fields and
to make new findings.
CRediT authorship contribution statement
Hodei Uzkeda: Conceptualization, Methodology, Formal analysis,
Data curation, Writing - original draft, Visualization. Mayte Bulnes:
Conceptualization, Formal analysis, Writing - review & editing,
Visualization, Funding acquisition. Josep Poblet: Conceptualization,
Writing - original draft, Visualization, Project administration, Funding
acquisition. Gonzalo Zamora: Conceptualization, Resources, Writing review & editing, Funding acquisition.
11. Conclusions
Acknowledgements
Two domains have been identified from the tectono-stratigraphic
point of view in the Boomerang Hills. In the southern domain, which is
the buried frontal part of the Andean Cordillera, the structural style
corresponds mostly to thin-skinned contractional tectonics, the structures are mainly Andean (Late Cenozoic) folds and faults, and the
stratigraphic succession is thick and almost complete including five
mega-sequences separated by unconformities: a PrecambrianOrdovician basement and four cover mega-sequences of SilurianDevonian, Carboniferous, Mesozoic and Cenozoic age. In the northern
domain, which belongs to the portion of the Beni-Chaco Plain adjacent
to the Andes, the structural style is mostly thick-skinned of extensional
type, the structures are mainly pre-Andean (Cambrian to Early
Cenozoic) faults, and the stratigraphic sequence is thinner because
Silurian-Carboniferous rocks are almost missing.
The three most striking structural events occurred in the Boomerang
Hills are: 1) formation of growth normal faults during three Pre-Andean
extensional events of Cambrian-Ordovician, Silurian, and Jurassic-Early
Cretaceous age, and normal reactivation of previous faults during early
Cenozoic (Andean); 2) flexure of the south part of the basement due to
quick subsidence during the Silurian-Devonian and slower during the
Carboniferous and possibly Permian-Triassic, so that the dipping portion of the basement became the southern slope of a high and behave as
a gravity-driven, extensional detachment; and 3) reactivation of the
extensional detachment as the basal detachment of oblique thrusts and
folds developed during a transpressional event of late Cenozoic age
(Andean), as well as selective reactivation of inherited normal faults.
The largest amount of displacement took place in the form of shortening
during the Andean event, while the maximum strain values were
achieved during the Andean event and during the basement rotation of
Paleozoic age.
The Paleozoic configuration of this region exerted a strong influence
on the main features of the frontal part of the Andes in terms of types of
structures, and their orientation, position and size. Thus, the position of
the Andean basal detachment within the ductile Silurian rocks was
controlled by the mechanical stratigraphy of the Paleozoic succession,
and its orientation, geometry and dimensions replicated the orientation, shape and size of a pre-Andean extensional detachment/slope of a
basin high. The development of one or another type of frontal structure
(tectonic wedge or faults and folds) responded to the angular relationships, geometry and location of the Paleozoic cover rocks and the
underlying basement.
The workflow followed in this study, which incorporates geological
interpretation of 2D/3D seismic data and wells, construction of a
transect across the area by projecting adjacent geological interpretations, construction of a 3D model, geometric/kinematic and mechanical
restoration in 2D and 3D accompanied by decompaction, and construction of maps over time of different stratigraphic units derived from
the present-day 2D sections, from the 3D model restorations and from
H. Uzkeda has been supported by research contract CN-16-014
(“Convenio específico para la realización de un trabajo de investigación
post-doctoral en la disciplina de Geología”) under the framework
agreement between Repsol Exploración S.A. and the University of
Oviedo. We are grateful to Repsol Exploración S.A. for providing the
well and seismic data. We thank J. Flinch, M. Masini and B. Amigo from
Repsol Exploración S.A. for helping us with the data and contracts. We
would like to acknowledge financial support by research project CGL
2015-66997-R (“Aplicación del análisis del plegamiento a la
investigación de recursos geológicos” -AAPLIREGE-) funded by the
Spanish Ministry for Economy and Competitiveness and the European
Fund for Regional Development (FEDER), and research project FCGRUPIN-IDI/2018/000216 (“Ayudas a organismos públicos de
investigación para apoyar la actividad que desarrollen sus grupos de
investigación en el Principado de Asturias en el periodo 2018–2020”)
funded by the Principado de Asturias. We thank Midland Valley and
Petroleum Exploration Experts for permission to use the software
Move© (Academic site software license and support agreement number
1915/Contract reference Agr_ULA_UniOviedo_18_11_01), Schlumberger
for permission to use the software Petrel© and Dynel© (Academic
agreement 1-1C5GCE1 Revision 1), HIS Markit for permission to use the
software Kingdom Suite© (University Grant Program), and Richard
Allmendinger and Nestor Cardozo for permission to use the software
Stereonet. We also thank J.G. Antuña for maintaining the software for
seismic interpretation, restoration and modeling. Comments by the
Editor Bernard Colletta, as well as Patrice Baby and anonymous reviewers have substantially improved the final version of the manuscript.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.marpetgeo.2020.104299.
References
Anderson, R.B., Long, S.P., Horton, B.K., Calle, A.Z., Ramirez, V., 2017. Shortening and
structural architecture of the Andean foldthrust belt of southern Bolivia (21°S): implications for kinematic development and crustal thickening of the central Andes.
Geosphere 13 (2), 538–558.
Angevine, C.L., Heller, P.L., Paola, C., 1990. Quantitative sedimentary basin modelling.
AAPG Contin. Educ. Course Note Ser. 32, 1–161.
Baby, P., Herail, G., López, J.M., Oller, J., Pareja, J., Sempere, T., Tufiño, D., 1989.
Structure de la Zone subandine de Bolivie: influence de la géométrie des séries
sédimentaires antéorogéniques sur la propagation des chevauchements. Comptes
Rendus de l'Académie des Sciences. Série 2 309 (17), 1717–1722.
Baby, P., Moretti, I., Guillier, B., Limachi, R., Mendez, E., Oller, J., Specht, M., 1995.
Petroleum systems of the northern and central Bolivian sub-Andean zone. In:
Tankard, A.J., Suárez S, R., Welsink, H.J. (Eds.), Petroleum Basins of South America.
vol. 62. AAPG Memoir, pp. 445–458.
20
Marine and Petroleum Geology 116 (2020) 104299
H. Uzkeda, et al.
Monos – Machareti (!) Petroleum System. US Geological Survey Open-File Report, pp.
16 99-50-C.
Maerten, F., 2010. Geomechanics to Solve Geological Structure Issues: Forward, Inverse
and Restoration Modeling. PhD thesis. Université Montpellier II, pp. 455.
Maerten, F., Maerten, L., 2001. Dynel: a new 3-D geomechanical method for modeling
geological structures. Stanford Rock Fract. Proj. 12, E1–E10.
Mercier, J.L., Sebrier, M., Lavenu, A., Cabrera, J., Bellier, O., Dumont, J.-F., Machare, J.,
1992. Changes in the tectonic regime above a subduction zone of Andean type: the
Andes of Peru and Bolivia during the Pliocene-Pleistocene. J. Geophys. Res. 97 (B8),
11945–11982.
Mescua, J.F., Giambiagi, L.B., 2012. Fault inversion vs. new thrust generation: a case
study in the Malargüe fold-and-thrust belt, Andes of Argentina. J. Struct. Geol. 35,
51–63.
Mingramm, A., Russo, A., Pozzo, A., Cazau, L., 1979. Sierra Subandinas. In: Proceedings
of the Segundo Simposio Geología Regional Argentino, Córdoba, Argentina, pp.
95–138.
Moretti, I., Baby, P., Méndez, E., Zubieta, D., 1996. Hydrocarbon generation in relation to
thrusting in the Subandean zone from 18° to 22°S, south Bolivia. Petrol. Geosci. 2,
17–28.
Moretti, I., Labaume, P., Sheppard, S.M.F., Boulègue, J., 2002. Compartmentalisation of
fluid migration pathways in the sub-Andean Zone, Bolivia. Tectonophysics 348, 5–24.
Nürnberg, D., Müller, R.D., 1991. The tectonic evolution of the south Atlantic from late
jurassic to present. Tectonophysics 191 (1–2), 27–53.
Perez, N.D., Horton, B.K., Carlotto, V., 2016. Structural inheritance and selective reactivation in the central Andes: Cenozoic deformation guided by pre-Andean structures in southern Peru. Tectonophysics 671, 264–280.
Sempere, T., 1994. Kimmeridgian? to Paleocene tectonic evolution of Bolivia. In: Salfity,
J.A. (Ed.), Cretaceous Tectonics in the Andes: Earth Evolution Sciences Monograph
Series. Vieweg Publications, Wiesbaden, pp. 168–212.
Sempere, T., 1995. Phanerozoic evolution of Bolivia and adjacent regions. In: In: Tankard,
A.J., Suárez S, R., Welsink, H.J. (Eds.), Petroleum Basins of South America, vol. 62.
AAPG Memoir, pp. 207–230.
Sempere, T., Baby, P., Oller, J., Hérail, G., 1991. La napa de Calazaya: una prueba de
acortamientos importantes controlados por elementos paleoestructurales en los
Andes bolivianos. Revista Técnica de YPFB 12, 229–234.
Sempere, T., Herail, G., Oller, J., Bonhomme, M.G., 1990. Late Oligocene-Early Miocene
major tectonic crisis and related basins in Bolivia. Geology 18, 946–949.
Suppe, J., Connors, C.D., Zhang, Y., 2004. Shear fault-bend folding. In: In: McClay, K.R.
(Ed.), Thrust Tectonics and Hydrocarbon Systems, vol. 82. AAPG Memoir, pp.
303–323.
Tankard, A.J., Uliana, M.A., Welsink, H.J., Ramos, V.A., Turic, M., França, A.B., Milani,
E.J., De Brito Neves, B.B., Eyles, N., Skarmeta, J., Santa Ana, H., Wiens, F., Cirbián,
M., López Paulsen, O., Germs, G.J.B., De Wit, M.J., Machacha, T., Miller, R.M.C.G.,
1995. Structural and tectonic controls of basin evolution in Southwestern Gondwana
during the Phanerozoic. In: In: Tankard, A.J., Suárez S, R., Welsink, H.J. (Eds.),
Petroleum Basins of South America, vol. 62. AAPG Memoir, pp. 5–52.
Uba, C.E., Heubeck, C., Hulka, C., 2006. Evolution of the late Cenozoic Chaco foreland
basin, southern Bolivia. Basin Res. 18, 145–170.
Uba, C.E., Kley, J., Strecker, M.R., Schmitt, A.K., 2009. Unsteady evolution of the Bolivian
Subandean thrust belt: the role of enhanced erosion and clastic wedge progradation.
Earth Planet Sci. Lett. 281, 134–146.
Welsink, H.J., Franco, M.A., Oviedo, G.C., 1995a. Andean and pre-Andean deformation,
Boomerang Hills area, Bolivia. In: In: Tankard, A.J., Suárez S, R., Welsink, H.J. (Eds.),
Petroleum Basins of South America, vol. 62. AAPG Memoir, pp. 481–499.
Welsink, H.J., Martinez, E., Aranibar, O., Jarandilla, J., 1995b. Structural inversion of a
Cretaceous rift basin, southern Altiplano, Bolivia. In: In: Tankard, A.J., Suárez S, R.,
Welsink, H.J. (Eds.), Petroleum Basins of South America, vol. 62. AAPG Memoir, pp.
305–324.
Baby, P., Specht, M., Oller, J., Montemurro, G., Colletta, B., Letouzey, J., 1994. The
Boomerang-Chaparé transfer zone: structural interpretation and experimental approach. In: Roure, F., Ellouz, N., Shein, V.S., Skvortsov, I. (Eds.), Geodynamic
Evolution of Sedimentary Basins, International Symposium, Moscow, Russia, pp.
203–218.
Bally, A.W., Snelson, S., 1980. Realms of subsidence. In: In: Miall, A.D. (Ed.), Facts and
Principles of World Petroleum Occurrence, vol. 6. Canadian Society of Petroleum
Geologists Memoir, pp. 9–94.
Bevis, M., Kendrick, E., Smalley JR, R., Brooks, B., Allmendinger, R., Isacks, B., 2001. On
the strength of interpolate coupling and the rate of back arc convergence in the
central Andes: an analysis of the interseismic velocity field. G-cubed 2
(2001GC000198), 1–16.
Bond, G.C., Kominz, M.A., Devlin, W.J., 1983. Thermal subsidence and eustasy in the
Lower Paleozoic miogeocline of western North America. Nature 306, 775–779.
Carey, S.W., 1958. The orocline concept in geotectonics. Proc. Royal Soc. Tasmania 89,
255–258.
Dunn, J.F., Hartshorn, K.G., Hartshorn, P.W., 1995. Structural styles and hydrocarbon
potential of the sub-Andean thrust belt of southern Bolivia. In: In: Tankard, A.J.,
Suárez S, R., Welsink, H.J. (Eds.), Petroleum Basins of South America, vol. 62. AAPG
Memoir, pp. 523–543.
Geobolivia, 2000. Mapa Geológico de Bolivia. escala 1:1 000 000.
Giraudo, R., Limachi, R., 2001. Pre-Silurian control in the genesis of the central and
southern Bolivian fold belt. J. S. Am. Earth Sci. 14 (7), 665–680.
Gubbels, T.L., Isacks, B.L., Farrar, E., 1993. High-level surfaces, plateau uplift, and
foreland development, Bolivian Central Andes. Geology 21, 695–698.
Hall, S.A., Bird, D.E., Mclean, D.J., Towle, P.J., Grant, J.V., Danque, H.A., 2018. New
constraints on the age of the opening of the South Atlantic basin. Mar. Petrol. Geol.
95, 50–66.
Hindle, D., Kley, J., Klosko, E., Stein, S., Dixon, T., Norabuena, E., 2002. Consistency of
geologic and geodetic displacements during Andean orogenesis. Geophys. Res. Lett.
29 (8) 29-31 – 29-34.
Hinsch, R., Krawczyk, C.M., Gaedicke, C., Giraudo, R., Demuro, D., 2002. Basement
control on oblique thrust sheet evolution: seismic imaging of the active deformation
front of the Central Andes in Bolivia. Tectonophysics 355, 23–39.
Hinsch, R., Krawczyk, C.M., Andina-Boomerang Team,Giraudo, R., Demuro, D., Rebay,
G., Goitia, V.H., 2003. Structural modelling and deformation balancing in the
foothills of the Bolivian Andes: constraints on accretion kinematics in curved thrust
belts. Revista Técnica de YPFB 21, 117–124.
Husson, L., Moretti, I., 2002. Thermal regime of fold and thrust belts–an application to the
Bolivian sub Andean zone. Tectonophysics 345, 253–280.
Illich, H.A., Haney, F.R., Mendoza, M., 1981. Geochemistry of oil from Santa Cruz basin,
Bolivia: case study of migration-fractionation. AAPG (Am. Assoc. Pet. Geol.) Bull. 65
(11), 2388–2402.
Isacks, B.L., 1988. Uplift of the central Andean plateau and bending of the Bolivian
orocline. J. Geophys. Res.: Solid Earth 93 (B4), 3211–3231.
Kley, J., 1996. Transition from basement-involved to thin-skinned thrusting in the
Cordillera Oriental of southern Bolivia. Tectonics 15 (4), 763–775.
Kley, J., 1999. Geologic and geometric constraints on a kinematic model of the Bolivian
orocline. J. S. Am. Earth Sci. 12, 221–235.
Klosko, E.R., Stein, S., Hindle, D., Kley, J., Norabuena, E., Dixon, T., Liu, M., 2002.
Comparison of GPS; seismological, and geological observations of Andean mountain
building. Plate Bound. Zones 30, 123–133.
Laffite, G., Del Vo, S., Aguilera, E., Goitia, V., Rebay, G., Lanussol, D., 1998. The
Petroleum Systems of the Boomerang Area. AAPG Search and Discovery, Bolivia
Article #90933.
Lease, R.O., Ehler, T.A., Enkelmann, E., 2016. Large along-strike variations in the onset of
Subandean exhumation: implications for Central Andean orogenic growth. Earth
Planet Sci. Lett. 451, 62–76.
Lindquist, S.J., 1998. The Santa Cruz – Tarija Province of Central South America: Los
21
Descargar